aad algorithms

[hackish 8]

I'm not a patent lawyer but there are a number of examples where algorithms have been patented and enforced over the years. In the computer world, check out the RSA algorithim (encryption) or LZW (data compression). One prime example was the .gif file format where many programmers were successfully forced to pay royalties for their software that used the LZW compression to create .GIF files.

MEMS pressure sensors can be extremely accurate. I have one that I've repeatedly used to show a 2' pressure change (going up and down stairs). Of course it was also highly affected by changes in barometric pressure.

Despite the costs there are advantages to requiring service every 4 years. Let's face it, we skydivers won't send our vigils in for checkup/maintenance if they don't require it. Hardware and algorithms can and have been updated. If you recall an issue a few years back where a factory was told that their batch of pressure sensors may have been faulty the possibility then exists to swap them out next 4 year maintenance without ever having to inconvenience the jumpers.

It is definitely possible to get the algorithm out of any AAD. That's what I do for a living, reverse engineer firmware. It's only a matter of time and money... lots of it.

One could definitely work with rate of change to determine if the AAD has truly entered freefall or if they "accelerated at an impossible rate". Don't forget the longer you wait to make the decision the more chance that the user is going to die should they really have exited the aircraft.

Is anyone aware of a fatality where the jumper left below activation altitude and their Cypres didn't fire? How about a save on a Vigil in the same circumstances? It's a tough argument to make either way.

-Michael

[JackC1 0]

Quote

I'm not a patent lawyer but there are a number of examples where algorithms have been patented and enforced over the years. In the computer world, check out the RSA algorithim (encryption) or LZW (data compression). One prime example was the .gif file format where many programmers were successfully forced to pay royalties for their software that used the LZW compression to create .GIF files.

I looked up the RSA and GIF US patents and they're for apparatus, not algorithms. So the algorithm is fair game so long as you don't use it on a computer or other suitable algorithm running apparatus. You could argue that this is an algorithm patent in all but name and I might agree with you, but it would cost you an arm and a leg to push the idea in court; and that is what a patent is there for. They're the nuclear deterrent of the corporate world.

[hackish 8]

I mentioned the LZW thing because I have a C&D from it many years ago. The patent itself is titled "Data Compression Method". It's been a lot of years since I looked at it but having been through the patent docs a number of times it definitely described the method (algorithm) not an apparatus. I got the C&D because I implemented it in assembly language and built one of the fastest GIF decoders at the time (yay 8086 processors). Patent 4814746 even had some source code in it. I'm not a patent lawyer but from my point of view algorithms can definitely be patented.

-Michael

[pchapman 261]

Just a couple quick points:

If someone is wondering what a Cypres measures, it would be worth opening up an expired Cypres 1. I did that once and posted pictures, but I didn't go through every chip's markings to try to determine what each did.

Vigils, for better or worse, do tend to be more 'sensitive'. A rigger I know put a rig in a heavy duty plastic bag, twisted the mouth of the bag, then squeezed the bag to get some air out of it. Pop went the Vigil, which he had forgotten was still turned on. One way or another the Vigil thought it was airborne and ready to fire, and then fired when the air pressure suddenly went up near ground level pressure.(But he didn't run the same test with a Cypres, so it is actually unknown what it would do.)

[nigel99 143]

Quote

Just a couple quick points:

If someone is wondering what a Cypres measures, it would be worth opening up an expired Cypres 1. I did that once and posted pictures, but I didn't go through every chip's markings to try to determine what each did.

Vigils, for better or worse, do tend to be more 'sensitive'. A rigger I know put a rig in a heavy duty plastic bag, twisted the mouth of the bag, then squeezed the bag to get some air out of it. Pop went the Vigil, which he had forgotten was still turned on. One way or another the Vigil thought it was airborne and ready to fire, and then fired when the air pressure suddenly went up near ground level pressure.(But he didn't run the same test with a Cypres, so it is actually unknown what it would do.)

I'd love to look under the hood of a Cypres.

Interesting about the plastic bag, pretty obvious that it could cause a problem (in hindsight of course).

Experienced jumper - someone who has made mistakes more often than I have and lived.

[JohnRich 4]

Quote

A rigger I know put a rig in a heavy duty plastic bag, twisted the mouth of the bag, then squeezed the bag to get some air out of it. Pop went the Vigil...

That's an old trick from decades back in the days of SSE Sentinel's, to check their functionality during rapidly rising air pressure.

[sundevil777 93]

Quote

One way or another the Vigil thought it was airborne and ready to fire, and then fired when the air pressure suddenly went up near ground level pressure.(But he didn't run the same test with a Cypres, so it is actually unknown what it would do.)

I believe the Vigil doesn't need to think it is airborne to fire, as confirmed by their warning that it might fire from closing an auto's trunk.

Quote

But he didn't run the same test with a Cypres, so it is actually unknown what it would do.

Since we know that the design has features to prevent that sort of activation, I wouldn't say it is "unknown what it would do."

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[davelepka 4]

Quote

I believe the Vigil doesn't need to think it is airborne to fire, as confirmed by their warning that it might fire from closing an auto's trunk

So it can fire in a plane if a door or window is opened, they can fire when you close the trunk of your car, and they can fire in a plastic bag. What I want to know is what the justification for progamming the device to make these events possible? What real-world scenario do they anticipate that would make it neccesary for the device to fire under these circumstances?

Short of any such explanation, why isn't the device programmed with some 'fail safes' in places to prevent such things?

Why would an AAD arm unless is has recognized a climb and reached an altitude of at least 500 or 600 ft? That would be the bare minnimum you could expect a jumper to be able to exit, get up to AAD firing speed, and still have time to get a reserve out (just barely), so what's the point in making an AAD that will fire without meeting those requirements?

I just don't get it.

[pchapman 261]

Regarding the sensitivity and 'arming' of the Vigil, I checked the manual, where it states:

Quote

Once your Vigil® is on, it will at each take-off (in max. 32 sec, from +150ft or 46m) switch to active mode.

I think what it is saying is that a climb of 150 ft within 32 seconds is considered a takeoff, which then 'arms' the Vigil. AAD's term is 'switching to active mode'.

This 150 ft active mode thing can also be the cause of Vigils not switching off all week if you leave it in a car and go up in elevation, not returning to the zero level.

I'm wondering if some of the car door type incidents occur after the car has driven somewhere higher -- So that all is needed is a sharp pressure increase to fire the Vigil, rather than some oscillating fluctuation of lower pressure followed by rapidly increasing pressure.

I'm not sure where into all of this one fits Vigil's "patented “permanent Left over Time Calculation” method." Whatever filtering or prediction it is doing, some spurious pressure oscillations get by it.

The Vigil on Pro mode will fire anywhere from 840 ft down to 150 ft, again from the manual. But with the belly to earth burble compensation of +260 ft for firing, I guess that means the Vigil is looking for a pressure corresponding to 410 ft as the lowest at which it will fire.

If that's right, it would imply that before firing, the pressure would not only have to drop 150 ft worth to go into active mode, but get up to 410 ft worth before a rapid pressure increase that actually sets the Vigil off.